Mycobacterium tuberculosis and M. avium commonly cause disseminated infections in patients with Acquired Immunodeficiency Syndrome (AIDS) leading to increased morbidity and mortality. Although M. tuberculosis infections can be treated effectively with existing drugs, increasing frequency of infections with drug resistant M. tuberculosis and the desire to shorten the chemotherapeutic treatment period warrant the search for more effective therapies. M. avium infections are virtually untreatable with existing antibiotics, but is unclear whether this antibiotic resistance results from the impermeability of the M. avium cell wall or the innate resistance of the drug targets to the antibiotics. The development of new therapies for treating mycobacterial disease will require basic research to clarify and define both the failures of existing anti- mycobacterial drugs and also to define new drug targets of these intracellular pathogens. The goal of this proposal is to apply a molecular genetic approach to identify and characterize existing and novel drug targets of M. avium and M. tuberculosis infections. The necessary genetic tools to study mycobacterial functions have already been developed. Using novel cloning vectors, both single copy E. coli-mycobacteria shuttle phasmids and multi-copy shuttle plasmids, recombinant-DNA has been introduced into M. smegmatis and BCG at high efficiencies. Other genetic tools, including a shuttle mutagenesis system to construct auxotrophic strains, mycobacterial libraries generated in E. coli, the demonstration of beta galactosidase as a mycobacterial reporter gene, and macrorestriction analysis of mycobacterial genomes, have opened the possibility for the analysis of M. avium genes encoding potential drug targets. By cloning, characterizing, and sequencing the M. tuberculosis and M. avium DNA gyrase genes, we plan to study the molecular mechanisms involved in resistance to fluoroquinolones. Dihydrofolate reductase and thymidylate synthetase also represent excellent targets for chemotherapeutic agents. Trimethoprim- resistant M. smegmatis mutants have been obtained which should facilitate the characterization of the dihydrofolate reductase genes for M. avium and M. tuberculosis. In addition, the genes responsible for the synthesis of M. avium glycopeptidolipids will be analyzed. M. smegmatis should serve as a useful cloning host since it has a comparatively short generation time and is non-pathogenic. Furthermore, genetic systems for the manipulation of M. avium strains will also be developed to determine the mechanisms of resistance to known anti-tuberculosis drugs.